ABSTRACT
The HIV epidemic continues to disproportionately affect men who have sex with men (MSM). The government has sponsored interventions and strategies to increase regular HIV testing among MSM but those charged with quantitatively evaluating the degree of regularity in HIV testing face several challenges. In addition to HIV testing, assessing the regularity of a sequence of events over time is relevant to other public health problems. In principle, any binary response model for longitudinal data could be used to evaluate regularity in the context, with approximately the same results as what our method will give. This chapter shows that this approach is justified by certain assumptions about the underlying event process. Additionally, the use of this approach allows for an understanding of the nature of this latent process, specifically in the form of a one-to-one transformation of model parameters into the unspecified baseline hazard of the process, evaluated at fixed points defined by the observation times of the study.
